Method
Four tiers. Seven refusals. One throughline. Read these before you read the desks.
The four-tier scale
UN bodies, ICJ, court judgments, state commissions of inquiry, peer-reviewed academic work.
Major international journalism (Reuters, BBC, FT, NYT, Guardian, Al Jazeera) and accredited human-rights research (HRW, Amnesty, ICG, PEARL, ITJP).
Memoirs, partisan or advocacy publications, secondary compilations. Cited only in support of Tier A/B.
State PR, unverified reporting, single-source claims without a paper trail. Not cited as primary support.
The seven refusals
- No intake. No survivor testimony, no complaints, no first-hand reports. Route to PEARL, ITJP, Mnemonic, OHCHR Special Procedures, or UK 999 / Refuge 0808 2000 247 in immediate danger.
- No naming. No serving personnel. No survivors. No perpetrators outside Tier-A indictments already published.
- No counts. No casualty totals produced here. We cite the body that published the count.
- No verdicts. We record attribution gaps; we do not close them.
- No HUMINT. No on-the-ground interviews with named individuals, no informant networks, no drones, no photography of personnel or checkpoints.
- No glorification. UK Terrorism Act 2000 §12. The LTTE is referenced only in historical and accountability context.
- No surveillance. Satellite imagery only at base-footprint resolution. Never used to identify individuals.
The throughline
Every desk must strengthen the self-determination case. Findings that look, at first, like they weaken it — child recruitment, the Karuna split, the 1990 Northern Muslim expulsion, the Rajiv Gandhi assassination — are still published. They are paired with the doctrinal frame that absorbs them: carence de souveraineté, the state's own conduct, and the individual-vs-organisational distinction in international responsibility law. The Laboratory is honest about the record; the record is strong enough to carry the honesty.